Mahr and Divorce Leila Salarpour G 1 Mahr and Divorce: An Islamic Marriage Concept and Its Effects on Intrahousehold Bargaining Power of Couples Leila Salarpour Goodarzi Binghamton University February 2018 Abstract This paper investigates how the recently (2013) reformed mahr system in Iran affects the intrahousehold bargaining power of couples. Mahr—a monetary gift promised by the groom to the bride at the time of marriage—is a mandatory part of prenuptial agreements in Islamic marriage contracts and plays a crucial role in divorce cases. Under Islamic law, a husband must pay mahr whenever his wife requests it or when he initiates a divorce, a right granted only to men. Women can request a divorce for exceptional reasons such as observed domestic violence, husband’s infertility, or second marriage without the wife’s permission. Lacking satisfactory cause, a woman usually must forfeit part of her mahr when she negotiates for her husband’s approval of her divorce request. The new law obligates husbands to pay a capped value of mahr, up to 110 gold coins (around $35k USD), with any additional payment conditional on men’s individual wealth. The theoretical model of Islamic marriage and divorce presented here shows how mahr affects and alters a couple’s choices and market outcomes. Divorce documents hand-collected in Shiraz, Iran, provided empirical evidence that supports the predictions of this model for divorces both before and after the policy change. Compared to the initial amount promised in male-initiated divorces, actual payment of mahr decreased substantially, whereas mahr paid in successful cases initiated by women showed no significant decrease. Length of the negotiation process did not change. JEL Codes: D86, D19, J4, J1, J16 Keywords: Islamic Marriage, Mahr, Contracts, Divorce, Gender, Intrahousehold Bargaining
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Mahr and Divorce Leila Salarpour G
1
Mahr and Divorce: An Islamic Marriage Concept and Its Effects
on Intrahousehold Bargaining Power of Couples
Leila Salarpour Goodarzi
Binghamton University
February 2018
Abstract
This paper investigates how the recently (2013) reformed mahr system in Iran affects the intrahousehold bargaining power of couples. Mahr—a monetary gift promised by the groom to the bride at the time of marriage—is a mandatory part of prenuptial agreements in Islamic marriage contracts and plays a crucial role in divorce cases. Under Islamic law, a husband must pay mahr whenever his wife requests it or when he initiates a divorce, a right granted only to men. Women can request a divorce for exceptional reasons such as observed domestic violence, husband’s infertility, or second marriage without the wife’s permission. Lacking satisfactory cause, a woman usually must forfeit part of her mahr when she negotiates for her husband’s approval of her divorce request. The new law obligates husbands to pay a capped value of mahr, up to 110 gold coins (around $35k USD), with any additional payment conditional on men’s individual wealth. The theoretical model of Islamic marriage and divorce presented here shows how mahr affects and alters a couple’s choices and market outcomes. Divorce documents hand-collected in Shiraz, Iran, provided empirical evidence that supports the predictions of this model for divorces both before and after the policy change. Compared to the initial amount promised in male-initiated divorces, actual payment of mahr decreased substantially, whereas mahr paid in successful cases initiated by women showed no significant decrease. Length of the negotiation process did not change.
This paper examines how mahr, a monetary gift promised by the groom to the bride at the time of marriage,
affects Islamic divorce negotiations in Iran. The unique structure of Islamic marriage and divorce is part of
an asymmetric, one-sided system that reserves the right to divorce for men. the financial promise of mahr
can secure some level of decision making within marriage and negotiating power in divorce for women.
Under this specific framework, the approach to divorce is different for men and women.
In February 2013, a new policy in Iran capped the maximum value of mahr at 110 gold coins (around $35k
USD). If the marriage contract exceeds this cap, the husband must pay the first 110 gold coins but any
remaining balance depends on his wealth and ability to pay. In contrast, the previous policy allowed women
to use legal force, including imprisonment, if husbands refused to pay the promised value of mahr.
Importantly, the new law removed these heavy penalties when mahr exceeded the cap and allowed husbands
to pay any remaining balance in installments.
Many articles have investigated intrahousehold relationships (i.e., bargaining power within households,
marriage, and divorce), and some have focused on theories of family and tried to develop models that
explain the decision-making process within marriage (Becker 1980, Mancer & Brown 1980). No-fault
divorce laws in the USA generated studies that examined how such changes might affect the intrahousehold
relationship when both husbands and wives can initiate divorce. However, the one sided right of divorce
only for men and in response, using mahr as a bargain tool by women separates the structure of Islamic
divorce from others. To explain this mechanism, I defined a model of collective bargaining that expands
previous theories of marriage by adding specifications for mahr and no-fault divorce rights for men. Hence,
this model uses joint utility to compare a couple’s options before and after the policy, and examines how
the new policy affects behavior and outcomes for both spouses.
Policymakers justified the new law under the premise that the high price of mahr threatened married life
and increased the number of men who were jailed because they could not pay the promised mahr. Diya
Organization of iran1, reported that around 3,000 men were imprisoned in 2012. Under the new policy, this
figure decreased to around 2,000 in 2013 but rose again to around 2,300 by 2015.
I used data (1,000 cases) hand-collected from four different offices of marriage and divorce in Shiraz, Iran,
to examine the outcomes of successful divorce cases before and after the policy change and compared them
with the predictions of the model. Optimally, the court is the source for information about couples who
filed for divorce and failed to receive permission. However, gaining access to Iranian divorce records is
1 Diya Organization of iran is an institution that works with financial prisoners such as cases who commit mahr-related crime
Mahr and Divorce Leila Salarpour G
3
extremely hard, especially from the courts. No previously collected data provided enough information about
mahr and divorce. The data presented here offer many specific details, such as initial mahr values and
payments, that can help when studying the relationship between mahr and divorce. Divorce documents
separate all cases into three general groups: (i) divorces initiated by women, (ii) divorces initiated by men
or those where the couple mutually agreed to a separation, and (iii) registered but unconsummated
marriages. This paper examines the first two groups.
Many common marriage institutions (e.g., dowry, or bride-price) have been studied in developing countries
such as India, Bangladesh, and Nepal. Mahr differs from bride-price (i.e., money paid by the groom to the
bride's family) because the bride receives the gift. Importantly, mahr affects the bargaining power of men
and women and plays a key role in Islamic divorce. Few other researchers addressed mahr in studies that
analyzed how mahr co-exists with dowry (Bianquis 1996, Carroll 1986a), and how it affects male and
female bargaining power within that framework (Chowdhury et al. 2017, Ambrus et al. 2010). Various
endogenous factors (e.g., education, wealth level of families, etc.) determine the relationship between mahr
and marriage (Chowdhury et al. 2017). By exogenously restricting mahr, the new policy provides a natural
experiment for observing how mahr value affects couples’ bargaining power within intrahousehold
relationships and in divorce cases.
The new cap only affects couples whose marriage contracts stipulate initial mahr that exceeds 110 gold
coins. Therefore, I used cases with initial mahr greater or less than 110 gold coins as the control group and
treatment group, respectively. Because mahr affects men and women oppositely, their approach to divorce
must be studied separately. Additionally, because the new policy links the cap to husbands’ wealth, the
level of effects varies according the relative value of mahr for the husband.
The theory used here assumes that the probability of divorce requests will decrease for women and increase
for men. However, the absence of information about unsuccessful cases or reasonable data on the pool of
marriages that might have been affected by the policy makes calculation of this effect impossible. Rather,
we examined how the new policy affects mahr payments and whether those effects affirm our model’s
prediction. Using a difference-in-differences approach, I examined the effects of the new policy on the final
percentage of mahr paid and the length of the divorce process seperately, based on which spouse initiates
the divorce.
When men initiate divorce, the new policy negatively affects the value ratio of mahr paid over mahr
promised. I found no substantial change in cases where women requested a divorce. This follows the
Mahr and Divorce Leila Salarpour G
4
prediction of the model, which anticipates an ambiguous effect on observable divorce cases requested by
women.
Indicating the value of mahr at the time of marriage is a common practice in most Muslim societies, which
do not regulate the type or amount of mahr. Rather, mahr is based on an agreement between a couple.
Family status, education, wealth level of each side, and even age difference may determine this agreed
amount. At different time periods, specific types of mahr are more popular, but those types can change due
to factors including acceptance of certain kinds of goods such as land or rapid changes in inflation. In Iran,
the most common forms of mahr include money, gold, or properties such as houses or land. In the past two
decades, high inflation rates made the official gold coin one of the most popular currencies for mahr2.
In most cases, the amount of mahr is decided under the supervision of both families. The agreed amount is
written into the prenuptial marriage contract and, like the rest of the contract, it is binding. If his wife asks
for her mahr at any point in the marriage, the husband must pay what he promised. In Islamic principles,
the importance of mahr is so high that a widow receives her mahr, or its equivalent, before her husband’s
heirs divide their inheritance.
Traditionally, couples do not consider transferring the mahr unless they consider ending their marriage. If
a woman asks for her mahr, her action is usually seen as a threat signal for divorce. Unlike many other
religions, Islam has always accepted divorce, even though it is not favored socially in many countries,
including Iran. The adverse impacts of divorce affect women more than men, as per societal norms. In
recent years, divorce has increased and consequently has become more acceptable (Figure 1), especially in
the major cities, but the social costs of divorce are far less for men compared to women. Figure 1 presents
the total number of marriage and divorce cases in Iran between 2005 and 2015.
Islamic divorce regulations define main differences in how men and women approach divorce. Based on
Islamic practice and Article 1133 of Iran’s civil codes, the unilateral right to divorce belongs to men
(Osanloo 2009). Some believe that mahr for women is the equivalent of this right. Women request a divorce
only in exceptional and compelling cases (e.g., observed domestic violence, the inability of the man to
financially support his family, addiction, absence of the husband for four full years, divorce right included
in prenuptial agreement, etc.). Aside from such exceptions, a woman must negotiate to gain her husband’s
approval for divorce. Arrangements may include compensating the husband for his approval by forfeiting
2 Tamam Seke Bahar Azadi. Each gold coin weighs 8.133 g (0.287 ounces), and the price of an Iranian gold coin is linked to the world price of the gold. Further, Iran prices its official currency according to the demand for gold within the country and the exchange rate between the Iranian riyal and other commonly used currencies in international trade, especially USD, Euro, etc.
Mahr and Divorce Leila Salarpour G
5
part or all of the promised mahr or, in some cases, paying their husbands an extra amount. If the man does
not accept the best offer that his wife is willing and able to provide, he must pay the full amount of mahr.
In doing so, man reserves his divorce rights and ends the divorce process.
Considering that Muslims account for 23 percent of the world’s population, the number of studies
investigating Islamic concepts is surprisingly small. Regarding marriage market and family structure,
various studies examined dowry and its consequences in India, Bangladesh, etc. (Becker 1981, Rao 1993,
Anderson 2003). However, very few papers have examined Islamic marriage contracts. The current study
uses the change in mahr policy as a natural experiment to fill part of this gap. There have been many debates
about the role of mahr in marriage, women’s bargaining power during marriage, and in case of divorce, as
a counter-argument to men’s unilateral divorce rights. It is rational to assume that increasing the amount of
mahr will increase women’s bargaining power with men, and it is necessary to recognize that the influence
of mahr depends on a man’s wealth. If the relative value of mahr is small compared to the husband’s
financial situation, it will detrimentally affect a woman’s bargaining power because the husband will not
consider mahr worthy of negotiation. By limiting the power of mahr and linking it to a man’s wealth, the
new policy artificially limits the market mechanism. This paper illustrates a divorce framework wherein
men and women consider their utility in both marriage and divorce and analyze the net outcome of mahr
before deciding to continue their marriage or file for divorce. It then uses the new policy to determine how
the new limit on the amount of mahr might change behavior and identify the outcome they will face in this
new environment.
Section I explains some of the related legal concepts and the history of family law in Iran. Section II
describes the data and provides some information about the details of cases. Section III covers the
framework and model of the divorce and Section IV explains the general divorce framework and the
identification strategy. Section V reports the result, and Section VI discusses some the processes and results
while highlighting the limitations.
I. Iran and Family Laws
Family laws in Iran have changed many times in the past few decades. In 1930, the Modernization Act
established the minimum legal age of marriage (15 and 18 years of age for girls and boys, respectively),
starting in 1935 (Momeni,1972). Lacking a legitimate request to the court, Article 1041 of the Iranian Civil
Code makes marriage below these ages illegal. Even then, girls and boys younger than 13 and 15 years,
respectively, could not marry (Aghajanian,2001). In practice, the law had a smaller impact in rural areas,
where marriage at less than 13 years of age was still a social norm.
Mahr and Divorce Leila Salarpour G
6
During the post-World War II period (1955–1979), Iran faced significant structural changes to its economy,
accompanied by social reforms such as women’s suffrage and political participation. In the following years,
new family laws affected household formation by altering the level of fertility and household growth
(Aghajanian,2001).
Only two weeks after the 1979 Revolution, Iran suspended the Family Protection Law (FPL) of 1967, which
had been revised in 1975. FPL had three primary objectives: procedural limitations to men’s unilateral right
to divorce, a restriction on polygamy, and a qualification of men’s presumptive rights to child custody
(Osanloo 2009). Based on FPL, husbands were required for the first time to start a formal procedure to
request a divorce. This process did not remove their unilateral divorce rights, but rather required mediation
to resolve their problems. These laws also raised the age of marriage to 18 and 20 years for women and
men, respectively. Questioning its compatibility with Islamic practice, Iran removed FLP from the new
civil laws after the revolution.
By October 1979, previously suspended family courts resumed their activities regarding disputed divorces
under the new name of the Special Civil Courts. When divorce consent was mutual, husbands were only
required to register a divorce in the presence of two male witnesses. After many complications caused by
changes in the divorce process and the lack of clear rules, the post-revolution Parliament formed a
committee on family matters in 1984 to propose remedies for judicial confusion about family laws (Paidar
1995). By 1989, a more uniform set of family laws included many previous FPL regulations. In 1992, the
Amendments to Divorce Regulations required husbands to attend court arbitration and obtain a letter of
noncompatibility (Osanloo 2009). Also in 1992, a new law allowed women to request post-divorce
maintenance as compensation for their work at home during marriage3. In 1999, women once again were
able to initiate divorce if their husband married a second wife without their permission.
Regulations regarding child guardianship were the subject of many controversial discussions. Based on the
Iranian Civil Act4 , the mother could have custody of the children until they reach the age of 2 and 7 years
for boys and girls, respectively, if the couple lacked a prior agreement. Afterward, guardianship moved to
the father (Ebrahimi 2005). In 1997, a new set of child custody laws granted custody to the mother if she
could show that her husband was unfit to care for children (Mir-Hosseini 2000). In 2002, the court entrusted
guardianship to the mother from birth to 7 years, without regard to gender, and then to the father for girls
3 Ojratol Masal 4 Article 1169
Mahr and Divorce Leila Salarpour G
7
aged 7–9 years and boys aged 7–15 years, at which point the children could choose their guardian (Ebrahimi
2005). In all instances, the father was responsible for child support.
Recently, debate over women rights resulted in some changes to the content of marriage contracts, including
pre-written conditions that give women more power in the marriage, including women’s right to work or
travel abroad at will. Although the contract provides such information to increase awareness about possible
grounds for agreement, many women are not aware of these possibilities, especially in rural areas. In more
traditional cultures, prenuptial options are not acceptable. Interestingly, many men prefer to accept higher
mahr than relinquish these rights to women. Also, some parts of Iran still use other culturally defined
agreements like bride-price5, which has decreased over time.
II. Data
The data presented here were collected at four local offices of marriage and divorce in various parts of
Shiraz, Iran. The best and most comprehensive resource for data related to marriage and divorce documents
in Iran is the National Organization for Civil Registration (NOCR). To register a divorce case, couples must
take a court permit to a local office, which records their information (i.e., type of divorce, permit date, etc.),
and reports the divorce to NOCR. In the past, NOCR lacked comprehensive information related to factors
such as child guardianship or amount of mahr promised or paid.
Because local offices restrict access to their data due to privacy concerns, unauthorized people cannot see
any of the documents. However, a team of lawyers who were working on another project in local offices
could access the documents, and they helped me collect 1,000 data points for this paper. The data, which
were drawn from offices located in one high-income area, two middle-income areas, and one low-income
area, contain no personal identifiers, and there is no opportunity for follow up. To prevent possible effects
of the policy change on marriage contracts, this paper only used data from marriage dates that preceded the
new policy. Divorce documents were organized according to divorce date. The data represent a cross-
section from specific months in 2006–2009 and 2011–2016. Due to recently unstable inflation rates, gold
has turned to the most popular currencies for mahr agreements. When mahr value is defined in terms of the
Iranian Rials, the court uses a CPI table to convert the numbers to present value. Thus, men pay the present
value of mahr rather than the nominal amount stated in the marriage contract. Following the same process,
this paper calculated the value of a gold coin based on the price of gold at the time of divorce (i.e., when
men must buy and pay the mahr). The new policy defines the cap in terms of gold coins. This specific way
5 The literal meaning of Shir Baha is the cost of breast feeding
Mahr and Divorce Leila Salarpour G
8
of defining the cap is the main reason this paper makes comparisons using the same measurements rather
than the Iranian Rials. Every divorce document in the data contained information on divorce-related factors
such as birth dates, total amount of mahr promised and mahr paid, alimony, guardianship of children, etc.
This study discusses two kinds of divorce:
(i) male-initiated, including mutual decision;
(ii) female-initiated; and
A third group (i.e., formally registered but unconsummated marriages) follows different rules and must
annihilate the marriage. In unconsummated marriages, women can get no more than half of the promised
mahr unless ordered by the court. Additionally, because marriage termination in the third group usually
happens very near the time of marriage, it caused a huge decrease in the sample after the policy. Finally,
there is no way to determine who requested annihilation of the marriage because either spouse can request
it and the legal term is the same for both spouses.
Table 1 shows some basic information for men and women, based on divorce type.
*Table 1*
Table 2 reports the number of observations based on divorce date, separated by divorce type (i.e., before or
after the policy), and over or under the 110 gold coins (treatment and control groups, respectively).
*Table 2*
Notably, marriage age for men and women was distributed differently in the sample. Figure 2 illustrates the
distribution of age at marriage for men and women separately, based on the divorce type.
* Figure 2*
The average age at marriage increases over time in the data. The marriages of most spouses who married
at very early ages were registered before Iran’s 1979 Revolution (30–40 years ago); consequently, divorce
happens at older ages. Most divorces in this group were initiated by men. For women, the special shape of
age distribution at the time of marriage can result from a trade-off between schooling and marriage at
specific ages. Compared to women, marriage age for men follows a normal distribution and starts at a later
age.
Mahr and Divorce Leila Salarpour G
9
Figure 3 shows a normal age distribution at time of divorce that is almost the same for both men and
women, although the age range is different.
* Figure 3 *
Age difference between spouses in the data sample decreased over time. Figure 4 shows that age differences
are smaller in recent marriages.
* Figure 4*
The data also provide self-reported information about spouses’ employment status and possibly measure
wealth. This paper examines whether independent income can affect women’s divorce outcomes and
increase their probability of having an acceptable outside option. Among 807 women who reported their
occupation, 439 were housewives and 368 were employed. When viewed according to which spouse
initiated the divorce, 153 women in male-initiated cases reported working and 233 were housewives. In
female-initiated cases, 215 women were employed and 206 were housewives, indicating a significantly
higher rate of employed women in this scenario.
Children are an important factor in both the marriage and divorce markets. Even when children are old
enough to choose their guardian, couples must consider the legal possibilities and their own
responsibilities in terms of both cost and child support before requesting a divorce. For girls and boys
younger than 7 years of age, guardianship is both the responsibility and the right of the mother. The father
then takes guardianship and responsibility for children until they reach 9 and 15 years of age for girls and
boys, respectively. In all instances, the father should pay child support unless negotiated otherwise.
Because spouses can negotiate their custody rights and even cede custody in return for other possible
divorce outcomes, mahr can actually help women as they negotiate post-divorce child custody.
III. The Model
I use a model of collective bargaining in this paper to explain how mahr affects the joint utility of husband
(H) and wife (W). In this model, both husband and wife face a set of possible goods with their budget, and
they decide how to allocate their market and nonmarket resources to these goods. They choose the optimum
level of consumption, which lies on their utility frontier line (i.e., the highest level of joint utility a couple
can achieve in cooperative marriage, based on their budget constraint). Their joint decision is based on the
following function:
Mahr and Divorce Leila Salarpour G
10
max 𝛽𝑈(𝑀) + (1 − 𝛽)𝑈(𝑀) s. t budget,
where 𝑈(𝑀) is the the utility of the wife from marriage, 𝑈(𝑀) is the utility of the husband from
marriage, β assigns the wife’s level of bargaining power in the household decision making, and 1- 𝛽 is the
bargaining power of the husband. Thus, 0 < 𝛽 < 1. The value of 𝛽 is determined by factors like their share
of value of nonincome earnings, the proportion of income each individual provides, and social norms (i.e.,
age, gender, education, etc.). Other important factors determining β in this model include mahr (µ), which
establishes β in the Islamic marriage model, and the outside option (i.e., the highest utility that husband or
wife can gain outside their marriage). The outside option might be the utility of remaining single after
divorce, remarriage, or any other possible option in the event of a divorce. I assumed that the spouses’
bargaining power is determined before marriage and does not change over time. Exceptional possiblities of
change in this value include a sudden shock (e.g., an exogenous change regarding which spouse earns the
family income, a sudden increase in the wealth of one partner, or any factor that changes the possible best
option they face outside the marriage) to any of the factors that determine barganing power. A unique 𝛽∗
represents a couple’s bargaining power. If the value of 𝛽 is not correctly determined, the couple can
renegotiate to shift the ratio to the correct level (i.e., 𝛽∗).
Mahr can be interpreted as a debt the husband must pay to the wife. It becomes the legal property of the
wife the moment she asks for it, and she can decide how to spend it. Therefore, mahr has a positive effect
on 𝛽 and a negetive effect on 1-β. Figure 5 shows how mahr shifts the bargaining power from husband to
wife and therefore is their final point of equilibrium on the frontier.
* Figure 5 *
In Figure 5, the x-axis represents the utility of the husband and the y-axis represents the utility of the wife.
The shape of the frontier line depends on the shape of the individual utility function and how it will be
affected if they change their bundle of consumption. Point A on the graph represents the optimum joint
utility that husband and wife face based on their bargaining power without mahr. When mahr is
incorporated, the relative bargaining power of women increases from 𝛽 to 𝛽 . Consequently, their joint
utility is represented by point B on the graph. This increase in the wife’s utility can be as large as the exact
monetary value of her mahr.
No-fault, or one-sided, divorce is a legal right that enables each spouse to exit the marriage at any point
with given conditions based on the law but without considering the cause or reason for divorce. The simple
fact that a person who has this right might not find his or her marriage desirable anymore is a sufficient
Mahr and Divorce Leila Salarpour G
11
ground for divorce. If the no-fault divorce option is available to both parties and the utility of either spouse
is less than their highest utility outside the marriage, they can either renegotiate the allocation of goods
within the household or exit the marriage. In other words, the outside option is the threat point for the
marriage. When no-fault divorce is a legal option, this utility acts as a reservation utility level.
Figure 6 shows a basic model without mahr where no-fault divorce is possible for both parties. 𝑇 on the
x- axis represents the best outside option for the husband, and 𝑇 on y-axis shows the best outside option
for the wife. For the husband, any utility level to the right of 𝑇 (i.e., area C or D) represents a higher utility
for marriage compared to divorce. If the joint utility reaches the left side (i.e., area A or B), the husband
should consider an option that moves him back to a positive net utility of marriage or choose divorce. This
is also true for the wife: she would prefer divorce over marriage if the joint utility is in area A or D.
* Figure 6 *
When the utility of marriage is in area A, where it is less than the threat point for both husband and wife,
divorce is the final solution. If utility is in area C, both spouses would prefer staying in the marriage. If the
equilibrium point is in area B, the husband would prefer divorce but the wife would have higher utility if
she remains in the marriage. Thus, the husband could either get a no-fault divorce or negotiate with his wife
if he believes their β does not represent their optimum bargaining power. Changing the value of β would
shift both spouses to a point on the utility frontier where the husband’s utility is at least as good as his threat
point level while the wife stays in a higher level of utility compared to her threat point. If their joint utility
is in area D, the no-fault divorce option allows the wife to choose divorce or the husband will agree to
decrease his utility level by transferring it to the wife until she prefers staying in the marriage.
In a simpler model (i.e., both spouses have a right to no-fault divorce and mahr is not present), both spouses
can exit the marriage at area B or D compared to the Islamic marriage model, where only men have the
right to divorce without mutual agreement (Area A). Thus, the husband can request a divorce in area B
without considering his wife’s utility, but the wife cannot request for a divorce in area D unless she
somehow gains the husband’s agreement. Therefore, mahr becomes an essential factor in providing an exit
option for women while making divorce costlier for men.
Table 6 shows all possible outcomes that a husband and wife might face. Based on their choice within the
marriage as well as their outside option with 0 value of mahr (μ), α represents the fraction of mahr that
husband is obligated to pay if they divorce and 1-α is the fraction of mahr that is forgiven (i.e., no longer
Mahr and Divorce Leila Salarpour G
12
due to the wife). The largest value of α equals 1 when the husband pays the full amount of the promised
mahr. The smallest possible value of α is 0 when the woman gives up all of her mahr.
*Table 6*
Thus, the joint utility of marriage considers mahr. If the wife believes that the couple’s joint utility does not
reflect the mahr correctly, she can demand separate payment of her mahr (Table 6, column 2). On the other
hand, the husband should consider the fact that he owes a debt to his wife, which he can choose to pay off.
At that point, the wife can decide how to spend this money. As long as the value of mahr is at least as great
as the shift of his utility to the left on the frontier line, the husband prefers to accept the switch in his
bargaining power rather than pay the actual amount of mahr. If he decides to request a divorce, he should
pay the amount of mahr debt (µ).
Based on institutional differences, husband and wife must take different actions when they study their
situation concerning divorce. The husband has the right to initiate ending the marriage, a one-sided action.
When he considers his options, he must confront the fact that he owes his wife some amount of mahr that
he does not need to pay as long as he is married. In the absence of any policy or interference, if he asks for
a divorce while his wife prefers staying married, he must pay the full amount of mahr aside from all other
conditions that he might need to fulfill to gain a divorce (Figure 7). This effect on his threat point can be
shown by subtracting the value of mahr from his outside option, which decreases the total value of divorce
and shifts his threat point (TH) to the left by the values of the mahr (µ). Thus, his total net utility of marriage
is 𝑈 − 𝑇 + 𝜇. If the result is negative, he will prefer divorce. After he pays the mahr p, the wife’s
outside option (TW) increases by the value of µ because she will receive this money plus whatever utility
she has outside of marriage. However, even a negative value of the wife’s total utility after divorce (𝑇 +
μ − 𝑈 ) will not affect the husband’s decision about divorce.
* Figure 7 *
Previously, men had to pay the full amount of mahr. A man who promised a value of mahr higher than his
ability to pay would go to jail if his wife requested the money and he couldn’t provide it independently or
from outside resources. This creates a much higher value for the mahr and shifts his threat point (𝑇 ) almost
to point 0, and his total outside option becomes 𝑇 − 𝜔 + 𝑗, a high negative value. The new policy
removed imprisonment for promised mahr values above 110 gold coins, thus withdrawing jail from the
total utility function in binding cases. It also limited payments above the cap to the man’s wealth (ω) rather
than actual amount of mahr (µ). In other words, the policy changed mahr value in the utility functions of
Mahr and Divorce Leila Salarpour G
13
both women and men to 𝑓 (μ|𝜔), i=W or 𝑓 (μ|𝜔) = max(𝐶, 𝜔) 𝑖𝑓 μ > 𝐶, respectively. When ω<C, jail
punishment is still possible if men cannot pay the cap value. In this scenario, the new law shifts 𝑇 to the
right when ≥ 𝐶, increasing the likelihood that men will find the divorce option appealing. Even when men
could provide the remaining mahr from outside options, such as family or friends, and jail was not a a
possible outcome before the policy, a man’s 𝑇 will shift to right if his wealth level is less than the promised
amount of mahr. Therefore, it is less costly for men to provide the required mahr because payment depends
only on his wealth level. This decreases the effect of mahr on the joint utility function of marriage, increases
the probability that the husband will request a divorce, and decreases his overall mahr payment compared
to the original promised value.
In the absence of the policy, when a woman considers her options she faces a different approach. If she
requests a divorce, she must offer to forfeit part of her mahr. Her two best options are (i) getting the mahr
value while still married if she thinks it is not correctly accounted for in the joint utility function, or (ii)
requesting a divorce and giving up some part or all of her mahr. The husband either agrees to a divorce and
pays the remaining mahr or rejects the offer and pays the full value of mahr. In this situation, a husband
who wants to remain in the marriage must pay all of µ. To decide if he should agree to divorce, he calculates
the total value of his outside option (𝑇 − 𝛼μ) and compares it with the value of being married while paying
for the full amount of mahr (𝑈 (𝑀) − μ). In other words, he looks at his total net value of being married:
𝑈 (𝑀) − 𝑇 − (1 − 𝛼)𝜇 (Figure 8). Thus, his comparison is now between his utility of being married,
𝑈 (𝑀), and his new outside option, 𝑇 + (1 − 𝛼)𝜇, which is to the right of his original threat point line,
𝑇 . As long as 𝑈 (𝑀) ≤ 𝑇 + (1 − 𝛼)𝜇, he would accept his wife’s offer and grant her a divorce. The
highest value of an outside option that wife can provide for her husband using mahr is when she gives up
all of her mahr and α=0. If even after this offer the husband determines that the utility of marriage is larger
than the total utility of divorce, 𝑈 (𝑀) > 𝑇 − 𝜇, he will not accept the offer. In this case, the wife’s
utility of marriage increases by the amount of µ that was not already calculated in the original bargaining
power of the couple. If µ was fully calculated in the bargaining power before the transaction by eliminating
this factor, in the long term their joint utility of marriage eventually will return to its original place. Even if
the wife still faces a higher outside option, she will not be able to exit the marriage after the husband pays
the full mahr value. The more significant the value of mahr, the higher the shift of the new outside option
for the husband. Since higher mahr value decreases the right side of 𝑈 (𝑀) > 𝑇 − 𝜇, it increases the
likelihood that the husband will find a level of (1-α) appealing. The wife then considers the largest α that
makes her husband indifferent between the outside option and staying married or, in other words, the level
of α that causes 𝑈 (𝑀) = 𝑇 + (1 − 𝛼)μ.
Mahr and Divorce Leila Salarpour G
14
* Figure 8*
A third option for the wife involves renegotiating her bargaining power within the household. This situation
can happen if the husband faces a high positive net utility of marriage and paying mahr is costly for him
compared to what he can give up to increase his wife’s utility of marriage. In other words, if the wife
believes her mahr was not calculated in the value of β and she is facing a lower utility of marriage compared
to her outside option, she can use her mahr and eventually both spouses might end up in area C, where they
both have a positive value of being married without an actual transfer of mahr. This is especially the case
when the husband is not able to pay the full value of mahr.
Without the new policy, a higher value of mahr increases the likelihood that the husband will agree to a
divorce. If he cannot pay the full amount of promised mahr, he will agree because the alternative is staying
in the marriage, forced to pay whatever ratio of mahr he can and going to jail. If negotiations fail and
husband ends up in jail, the joint utility function will move away from its optimum level because he will
not be able to provide income and participate in housework. This noncooperative situation is costly for both
parties. At the same time, if the husband accepts his wife’s offer, his area will shift far to the right, toward
infinity, because his gains from agreeing to divorce are the utility of his outside option, a part of mahr, and
his freedom. This is the opposite of going to jail and highly positive. Under the new policy, the maximum
amount of mahr that a wife can forgive is her husband’s wealth level rather than the µ value. For the
husband, when μ > 𝜔 the highest amount a wife can offer is 𝑈 (𝑀) = 𝑇 + 𝜔. Therefore, if the promised
value of mahr is above the cap, the distance between 𝑇 and the original threat point caused by mahr
decreases, reducing both α and the probability of 𝑇 + μ ≥ 𝑈 (𝑀) that assures the possibility of divorce
for women. Eliminating jail from possible punishment decreases the value of μ when the husband can’t
afford to pay the mahr even with outside resources.
If the marriage utility of both spouses is negative without considering mahr, both parties will mutually agree
to a divorce. The presence of mahr makes no difference in the sense that the best outcome for both parties
is divorce. However, in this framework it is a question of how much of the mahr will be transferred. If the
husband’s value of marriage is less than his threat point plus the entire amount of mahr, he will pay the full
µ. If 𝑇 − μ ≤ 𝑈 (𝑀) < 𝑇 , they will agree to a ratio of mahr based on their outside option and how much
their utility will increase after they divorce.
If both spouses prefer the divorce option, the husband should request a divorce. In this case, both husband
and wife must agree to exchange some amount of mahr (Figure 9). Since he has the power to initiate divorce,
the husband offers the wife only the amount that keeps the wife indifferent between being married or
Mahr and Divorce Leila Salarpour G
15
divorced. If their joint utility is a point in area A, the wife will accept the offer. If the joint utility function
is in area 𝐴 , the husband cannot divorce without his wife’s consent. In this situation, he should compensate
her up to the level where he still prefers divorce over marriage. If he pursued a no-fault divorce, he would
have to pay the full value of µ. In this sense, he still benefits from divorce because he avoids paying the full
value of mahr. Finally, if the equilibrium point of marriage is in area 𝐴′′, the wife will agree to a mutual
divorce if she receives the full amount of mahr. Like her husband, full payment will increase her preference
for divorce and also make it more appealing than marriage.
* Figure 9: *
IV. Framework and Identification Strategy
The new policy can affect the couples in two ways: (i) asking for a divorce and determining whether it is
feasible, or (ii) changing their divorce outcomes. Before the new policy, all cases were treated the same.
Social changes and financial situation in the market effect people in the same direction. Decrease in real
income in the society makes it harder for men to pay the requirments they need to pay such as mahr to break
the marriage contract. In these situation, number of people with higher values of mahr or other payments
decreases. Estimating the effect of the policy on number of divorce cases initiated by men and women is
notNow, marriages with promised value below 110 gold coins remain unaffected, while the upperbund for
negotiation in cases with a promised amount above 110 shift to a husband’s wealth level if he cannot afford
to pay the total amount of promised mahr. To determine how the policy affects the results of divorce cases,
I use difference in difference regression comparing control and treatment groups for all cases toegther and
Figure 4: Fitted Value of Age Difference (Based on Length of Marriage)
Table 3, Summary Information on Children All Requested by men Requested by
women Girls boys All Girls boys All Girls boys All
Number of families with 287 281 498 127 99 201 160 182 297 Total number of 309 297 606 136 102 238 173 195 368 Number of cases at least 1 child living with mother
285 138 396 126 51 166 159 87 230
Number living with mother 306 149 455 135 52 187 171 97 268 Ratio of children living with mothers over all
women Number of families with independent children
114 33 81
Number of families with no children
132 110 22
Note: Table provides the summary information of guardianship of children and number of families with independent children and without children from the total 797 cases that provide information about children.
Mahr and Divorce Leila Salarpour G
26
Table 4- Divorce outcomes
Utility
Marriage Marriage with
Renegotiation
Divorce
Wife No Mahr 𝑈 𝑈 𝑇
Mahr 𝑈 |μ 𝑈 + μ 𝑇 +αμ
Husband No mahr 𝑈 𝑈 𝑇
Mahr 𝑈 |μ 𝑈 − μ 𝑇 − 𝛼μ
Figure 5: Join Utility of Spouses and Optimum Point with Mahr
𝑈
𝑈
𝑈
VH(µ)
VW(µ)1 − β
β
1 − β′
β′
A
B
Mahr and Divorce Leila Salarpour G
27
Figure 6: Joint Utility of Spouses (No-Fault Option for Both)
Figure 7: When Husband Considers the Divorce Option
𝑈
𝑈𝑇
𝑇
B
D
C
Both prefer the outside option to their marriage, divorce
Wife prefers marriage and husband prefers the outside option (either divorce or renegotiation)
Husband prefers marriage and wife prefers the outside option (either divorce or renegotiation)
Both prefer being married to the outsideoption
A
𝑈
𝑈𝑇 -VH(µ)
𝑇 +𝑉𝑊(µ)
B
D
C
Both prefer the outside option to their marriage
Wife prefers marriage, Husband prefers the outside option, either divorce or renegotiation
Husband prefers marriage, he will not act on divorce
Both prefer being married to the outside option
𝑇
A
If husband ask for divorce outside option becomes preferable for wife
Mahr and Divorce Leila Salarpour G
28
Figure 8: When Wife Considers the Divorce Option
Figure 9: Mutual agreement for divorce
𝑈
𝑈𝑇 + (1 − α)μ
𝑇 + αμ
B
D
C
Both prefer the outside option to their marriage
Wife prefers marriage, She does not act on anything
Husband prefers marriage, wife preferrs the outside option, renegotiation or nothing will happen
Both prefer being married to the outside option
𝑇
A
µ
µShift in joint utility point caused by mahr transfer.B is the alternative option
𝐴ʹ
𝑈
𝑈𝑇 -µ
𝑇 + μ
B
D
C
Both prefer the outside option to their marriage
Wife prefers marriage, Husband prefers the outside option, Either divorce or renegotiation
Husband prefers marriage, he will not act on divorce
Both prefer being married to the outside option
𝑇
A
𝐴′
A’’
Mahr and Divorce Leila Salarpour G
29
Figure 10: Kernel Density of Promised & Paid Mahr
Mahr and Divorce Leila Salarpour G
30
Figure 11: Kernel Density of Paid Mahr Ratio
Mahr and Divorce Leila Salarpour G
31
Figure 12: Polynomial Regression of Log Mahr and Paid Values (Men)
Mahr and Divorce Leila Salarpour G
32
Figure 13: Polynomial Regression of Log Mahr and Paid Values (Women)
R-sq 0.38 0.38 0.39 0.39 0.73 0.73 Standard errors in parentheses
="* p<0.05 ** p<0.01 *** p<0.001"
Note: Comparisons are only for cases in which divorce was initiated by men. Comparison is between the control group, cases with initial value below 110 gold coins and treated group with above this value. Policy*OVER110 is the variable of interest. Clustering is based on the month of divorce.
Time fix effects Yes Yes Yes Yes Yes Yes Cluster No Yes No Yes No Yes
N 420 420 398 398 420 420 R-sq 0.45 0.45 0.46 0.46 0.53 0.53 Standard errors in parentheses ="* p<0.05 ** p<0.01 *** p<0.001"
Policy*OVER110 = variable of interest Note: Comparisons are only for female-initiated divorce cases. Comparison is between controls (initial value below 110 gold coins) and treatment (above 110 gold coins). Clustering is based on divorce date.
Mahr and Divorce Leila Salarpour G
35
Table 7- Effect on the negotiation process (1) (2) (3) (4) (5) (6) Divorce process
N 387 387 387 421 421 421 R-sq 0.14 0.14 0.14 0.06 0.06 0.06 Standard errors in parentheses
="* p<0.05 ** p<0.01 *** p<0.001"
Note: Column 1-3 looks at the effect of the policy on negotiation time in terms of month on male initiated cases. Column 4-6 calculates the effect on female initiated divorces.
Table 8: Divorce Cases Requested by Men Compared to Women
Note: The dependent variable is a dummy equal to 1 if divorce is requested by the man and 0 by the woman. Columns 2–4 are clustered based on the divorce month. Column 3 is logit and column 4 is a probit regression.
Mahr and Divorce Leila Salarpour G
36
Mahr and Divorce Leila Salarpour G
37
References:
Abbasi-Shavazi, M. J., & McDonald, P. (2008). Family change in Iran: Religion, revolution, and the state.
Abbasi-Shavazi, M. J., & McDonald, P. (2008). Family change in Iran: Religion, revolution and the state.
In R. Jayakody, A. Thornton, & W. Axinn (Eds.), International family change: Ideational perspectives (pp.
177–198). New York: Lawrence Erlbaum
Abbasi-Shavazi, M. J., McDonald, P., & Hosseini-Chavoshi, M. (2009). The fertility transition in Iran:
Revolution and reproduction. Dordrecht: Springer.
Ambrus, A., Field, E., & Torero, M. (2010). Muslim family law, prenuptial agreements and the emergence
of dowry in Bangladesh. The Quarterly Journal of Economics, 125(3), 1349-1397. Retrieved from
http://www.jstor.org/stable/27867512
Anderson, S. and Bidner, C. (2015). Property Rights over Marital Transfers. The Quarterly Journal of
Economics.
Amin, S. and Cain, M. (1995). The rise of dowry in Bangladesh. In R. D. G. Jones, J. Caldwell and R.
D’Souza (Eds.), The continuing demographic transition, Oxford: Oxford University Press.
Bahramitash, R., & Kazemipour, S. (2006). Myths and realities of the impact of Islam on women: Changing
marital status in Iran. Critique: Critical Middle Eastern Studies.
BECKER, G. S. (1981): “A Treatise on the Family,” NBER Books.
Becker, G. S. (1988). Family economics and macro behaviour. The American Economic Review.
Binmore, K.,Rubinstein, A., and Wolinsky, A. (1986). The Nash bargaining solution in economic modeling.
RAND Journal of Economics
Chowdhury, C, Mallick, D and Chowdhury P. (2017). Natural Shocks and Marriage Markets: Evolution of
Mehr and Dowry in Muslim Marriages
Ebrahimi, S. (2005). Child Custody "(Hizanat)" under Iranian Law: An Analytical Discussion. Family Law
Quarterly, 39(2), 459-476. Retrieved from http://www.jstor.org/stable/25740500
Mahr and Divorce Leila Salarpour G
38
Jones, G. W. (2007a). Trends in marriage and divorce in peninsular Malaysia. Population Studies.
Kamal, S. 2001. Her Unfearing Mind: Women and Muslim Laws in Bangladesh. Ain O Salish Kendro,
Dhaka, Bangladesh
Lundberg, S., & Pollak, R. (1994). Noncooperative Bargaining Models of Marriage. The American
Economic Review, 84(2), 132-137. Retrieved from http://www.jstor.org/stable/2117816
Manser, M., & Brown, M. (1980). Marriage and Household Decision-Making: A Bargaining Analysis.
International Economic Review, 21(1), 31-44. doi:10.2307/2526238
Mir-Hosseini, Z 2000 Marriage on Trail:A Stuy of Islamic Family Law 2nd ed. London: I.B. Tauris
Osanloo, A. (2009). The Politics of Women's Rights in Iran. PRINCETON; OXFORD: Princeton
University Press. Retrieved from http://www.jstor.org/stable/j.ctt7sqth
Paidar, p 1995 women and the political process in twentieth-century Iran. Cambridge University press.
Persson, P. (2015). Social Insurance and the Marriage Market
Rao, V. (1993). The rising price of husbands: A hedonic analysis of dowry increases in rural India. The
Journal of Political Economy
Rapoport, Y. 2000. Matrimonial Gifts in Early Islamic Egypt. Islamic Law and Society
Welchman,L.2000. Beyond the code: Muslim Family Law and the Shari’s Judiciary in the Palestinian West
Bank. Kluwer Law International, The Hague.
The National Organization for Civil Registration of Islamic Republic of Iran
Formal Newsletter of Islamic Republic of Iran, announcement of the new law: